Since the development of DSM-III in 1980, and through its subsequent revisions (DSM-III-R, DSM-IV), DSM criteria have become almost universally accepted as a basis for diagnosis in the mental health professions. Despite this broad acceptance, there is a paucity of empirical data available that support the highly specified diagnostic formations embodied in manual. The impact that adding, removing, or otherwise altering criteria may have on disorder prevalence, natural history, or associated impairment has not been well studied. There is little consensus on the issue of comorbidity and the extent to which this phenomenon represents misplaced diagnostic boundaries, responses to common antecedents, or causal linkages between disorders. Finally, and especially important to child psychiatry, there is relatively little information available about developmental differences in the expression and significance of DSM-defined disorders. The proposed project would compile and analyze a database of diagnostic and symptom data collected by 16 investigations of youth, all of which used a DSM- linked version of the DISC. The total database would have data on approximately 19,000 children and adolescents -- longitudinal data is available on approximately 14,000 of these. In addition to the specific analytic aims outlined in the proposal (examining the validity of diagnostic subtypes and specifiers; understanding diagnostic boundaries and comorbidity; justifying diagnostic thresholds; studying the appropriateness of individual symptom criteria; comparing dimensional and categorical definitions), the investigators will work closely with the DSM-V Task Force and others in order to identify and address the most important and pressing diagnostic and classification issues facing the field. While the immediate practical benefit of this project will be to inform the development of DSM-V, its longer-term importance will be to address the epistemological underpinnings of child and adolescent psychiatric nosology in general. In a cost-efficient, manner this project will afford a unique opportunity to address many questions that can only be answered with such a large aggregated dataset with well-defined samples assessed with a standardized, comprehensive and transparent methodology. The aggregated data will also represent an invaluable data resource for future research.